summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
* Avoid construction of intermediate queuebug24433Matthew Sackman2011-09-221-6/+7
|
* merge default into bug24433Matthias Radestock2011-09-191-116/+101
|\
| * count queue consumers as required by the specMatthias Radestock2011-09-191-10/+9
| |
| * merge default into bug24433Matthias Radestock2011-09-181-19/+13
| |\
| * \ merge default into bug24433Matthias Radestock2011-09-180-0/+0
| |\ \
| * | | improved compiling- and working-nessMatthias Radestock2011-09-181-34/+33
| | | |
| * | | merge from defaultMatthias Radestock2011-09-181-48/+47
| |\ \ \
| * | | | work in progressMatthias Radestock2011-09-171-140/+117
| | | | | | | | | | | | | | | | | | | | | | | | | This doesn't even compile; but I want to stash it away so I can cherry pick some of the refactorings and tweaks for 'default'.
* | | | | cosmeticMatthias Radestock2011-09-191-3/+2
| |_|_|/ |/| | |
* | | | a bit of inlining, tweaking and cosmeticsMatthias Radestock2011-09-181-18/+13
| |_|/ |/| |
* | | cosmeticMatthias Radestock2011-09-181-7/+5
| |/ |/|
* | extract ack handling logicMatthias Radestock2011-09-181-34/+33
| | | | | | | | in common between 'ack', 'reject' and 'requeue'
* | extract correlation between consumer_count and limiter registrationMatthias Radestock2011-09-171-23/+21
| | | | | | | | | | | | | | | | | | | | | | ...and fix a bug in the process: during consumer registration we were using the *old* cr, with the old (and usually undefined) limiter, to check whether the channel is blocked. Thus we would end up running through the message queue unnecessarily. No big deal, but certainly not what was intended. Also, make the update_ch_record return the record, which makes the function more composable.
* | rename maybe_store_ch_record to update_ch_record and drop assertionsMatthias Radestock2011-09-171-27/+20
|/ | | | | ...and some associated cosmetic shuffling and inlining. The assertions aren't all that useful and just clutter the code.
* cosmetic: slightly more sensible field orderMatthias Radestock2011-09-161-5/+5
|
* merge bug24428 into defaultMatthias Radestock2011-09-163-177/+214
|\
| * merge bug23596 into defaultMatthias Radestock2011-09-162-22/+27
| |\
| | * Merged bug24424 into defaultEmile Joubert2011-09-161-2/+5
| | |\
| | | * Whitespacebug24424Emile Joubert2011-09-161-2/+2
| | | |
| | | * Fix duplicated example para when generating rabbitmqctl docsJim Apperly2011-09-151-2/+5
| | | |
| | * | merge bug24425 into default (FHC can become bottleneck when large numbers of ↵Alexandru Scvortov2011-09-151-20/+22
| | |\ \ | | | |/ | | |/| | | | | durable queues first receive a msg)
| | | * Use ets table instead of dictbug24425Matthew Sackman2011-09-141-20/+22
| | | |
| * | | use 'sets' instead of 'gb_sets' for the 'blocking' setbug23596Matthias Radestock2011-09-161-10/+10
| | | | | | | | | | | | | | | | | | | | This set will generally be quite small, there is just one of them and we don't need ordering, and generally performance here isn't an issue.
| * | | some tweaks for sake of symmetryMatthias Radestock2011-09-161-16/+13
| | | | | | | | | | | | | | | | ...and some cosmetics
| * | | minor simplifying refactorMatthias Radestock2011-09-161-8/+5
| | | |
| * | | merge two foldsAlexandru Scvortov2011-09-161-7/+7
| | | |
| * | | cosmetic changes and some minor refactoringMatthias Radestock2011-09-161-93/+82
| | | |
| * | | rename consumer_monitors to queue_consumersAlexandru Scvortov2011-09-151-29/+32
| | | |
| * | | remove unnecessary checkAlexandru Scvortov2011-09-151-7/+3
| | | |
| * | | consumer_monitors is a dict againAlexandru Scvortov2011-09-151-22/+24
| | | |
| * | | less arbitrary division of labourAlexandru Scvortov2011-09-151-8/+9
| | | |
| * | | remove no-opAlexandru Scvortov2011-09-151-1/+0
| | | |
| * | | monitor_queue/demonitor_queue are symmetric; don't use process dictionaryAlexandru Scvortov2011-09-151-126/+149
| | | |
| * | | consumer_monitors is now a set of QPidsAlexandru Scvortov2011-09-151-19/+25
| | | |
| * | | DeMorgan a caseAlexandru Scvortov2011-09-141-5/+7
| | | |
| * | | merge from default + cosmeticAlexandru Scvortov2011-09-141-55/+72
| |\ \ \ | | |/ / | |/| |
| | * | refactorAlexandru Scvortov2011-09-131-13/+6
| | | |
| | * | merge flow and (consumer, confirm, starts) monitorsAlexandru Scvortov2011-09-131-17/+20
| | | |
| | * | merge consumer and (confirms, stats) monitorsAlexandru Scvortov2011-09-131-37/+37
| | | |
| | * | merge stats and confirm monitorsAlexandru Scvortov2011-09-131-24/+35
| | | |
| | * | remove unnecessary variableAlexandru Scvortov2011-09-131-51/+40
| | | |
| | * | merge default into bug23596Alexandru Scvortov2011-09-1311-46/+57
| | |\ \
| | * | | refactorAlexandru Scvortov2011-09-061-5/+7
| | | | | | | | | | | | | | | | | | | | And mark the queue as un-monitored when it goes DOWN.
| | * | | only monitor queues that have messages on themAlexandru Scvortov2011-09-061-35/+51
| | | | |
* | | | | And now it's like well beautiful. Wicked.bug24428Matthew Sackman2011-09-151-22/+17
| | | | |
* | | | | Scatter/Gather. Eliminates gaps between txns and thus goes much faster.Matthew Sackman2011-09-151-19/+30
| | | | |
* | | | | Rework binding recoveryMatthew Sackman2011-09-151-14/+21
|/ / / /
* | | | fix a specMatthias Radestock2011-09-141-1/+1
| |_|/ |/| | | | | | | | that was causing dialyzer to complain, rightly.
* | | Merge heads.Jerry Kuch2011-09-131-11/+7
|\ \ \
| * \ \ Merging headsMatthew Sackman2011-09-131-11/+7
| |\ \ \ | | |_|/ | |/| |
View Lorry Controller Status